With the current rawhide version of oprofile, binutils and glibc opreport segfaults when using the -l option (see below)

I saw in koji that the rebuild of oprofile for F11 failed.

The most current working version oprofile-0.9.4-3.fc10 has the same problem as described in bug #467651
Recompiling from source (as suggested in comment #7) didn't help a lot as opreport is segfaulting now (see below)

binutils-2.19.51.0.2-16.fc11.i586
binutils-devel-2.19.51.0.2-16.fc11.i586
glibc-2.9.90-8.1.i686
glibc-devel-2.9.90-8.1.i586


gdb --args opreport -l --merge tgid
...
...
Counted CPU_CLK_UNHALTED events (Unhalted clock cycles) with a unit mask of 0x00 (Unhalted core cycles) count 100000
Samples on CPU 0
Samples on CPU 1

Program received signal SIGSEGV, Segmentation fault.
interesting_symbol (sym=0x0) at bfd_support.cpp:336
336		if (!sym->section) {
Missing separate debuginfos, use: debuginfo-install libgcc-4.4.0-0.23.i586 libstdc++-4.4.0-0.23.i586
(gdb) bt
#0  interesting_symbol (sym=0x0) at bfd_support.cpp:336
#1  0x080abd01 in op_bfd::get_symbols (this=0xbffff234, symbols=@0xbffff194) at op_bfd.cpp:217
#2  0x080acb03 in op_bfd (this=0xbffff234, fname=@0x81ac0d8, symbol_filter=@0x8198384, extra_images=@0xbffff4cc, ok=@0xbffff31f) at op_bfd.cpp:171
#3  0x08084b9d in populate_for_image (samples=@0xbffff4a8, ip=@0x81ac0d8, symbol_filter=@0x8198384, has_debug_info=0x0) at populate.cpp:70
#4  0x080544e9 in opreport (spec=@0xbffff634) at opreport.cpp:576
#5  0x0805b7c1 in run_pp_tool (argc=4, argv=0xbffff734, fct=0x8053d50 <opreport>) at common_option.cpp:210
#6  0x0804cabd in main (argc=4, argv=0xbffff734) at opreport.cpp:590